Samuel’s Part

Data Import & Prep

# importing frisk data
url <- "https://www.nyclu.org/sites/default/files/field_documents/2018_sqf_database.xlsx"
destfile <- "X2018_sqf_database.xlsx"
curl::curl_download(url, destfile)
SQF2018 <- read_excel(destfile) %>%
  mutate(SUSPECT_HEIGHT = as.numeric(SUSPECT_HEIGHT))

url <- "https://www1.nyc.gov/assets/nypd/downloads/excel/analysis_and_planning/stop-question-frisk/sqf-2017.xlsx"
destfile <- "sqf-2017.xlsx"
curl::curl_download(url, destfile)
SQF2017 <- read_excel(destfile)

SQF1718 <- SQF2018 %>%
  mutate(STOP_FRISK_ID = STOP_FRISK_ID + nrow(SQF2017)) %>%
  rename(STOP_FRISK_TIME = "Stop Frisk Time") %>%
  full_join(SQF2017)

SQF1718 <- SQF1718[order(SQF1718$STOP_FRISK_ID),]

# importing codebook
url <- "https://www.nyclu.org/sites/default/files/field_documents/2018_sqf_codebook.xlsx"
destfile <- "X2018_sqf_codebook.xlsx"
curl::curl_download(url, destfile)
SQF2018_cb <- read_excel(destfile)

# write to csv file
#write_csv(SQF1718, "/Users/slee19/Documents/Junior\ First\ Semester/SQF1718.csv")

View(SQF2018_cb)

SQF1718 <- read_csv("https://github.com/samuellee19/CS5_Project/raw/master/SQF1718.csv")

## Parsed with column specification:
## cols(
##   .default = col_character(),
##   STOP_FRISK_ID = col_double(),
##   STOP_FRISK_DATE = col_datetime(format = ""),
##   STOP_FRISK_TIME = col_datetime(format = ""),
##   YEAR2 = col_double(),
##   ISSUING_OFFICER_COMMAND_CODE = col_double(),
##   SUPERVISING_OFFICER_COMMAND_CODE = col_double(),
##   OBSERVED_DURATION_MINUTES = col_double(),
##   STOP_DURATION_MINUTES = col_double(),
##   SUSPECT_HEIGHT = col_double(),
##   STOP_LOCATION_PRECINCT = col_double(),
##   STOP_LOCATION_X = col_double(),
##   STOP_LOCATION_Y = col_double()
## )

## See spec(...) for full column specifications.

# make data.frame a spatial object
test <- SQF1718[!is.na(SQF1718$STOP_LOCATION_X),]
test <- test[!is.na(test$STOP_LOCATION_Y),]

coordinates(test) <- ~STOP_LOCATION_X + STOP_LOCATION_Y


test_coords <- as_tibble(coordinates(test))

# SET the CRS of the object
proj4string(test) <- CRS('+init=epsg:2263 +proj=lcc +lat_1=41.03333333333333 +lat_2=40.66666666666666 +lat_0=40.16666666666666 +lon_0=-74 +x_0=300000.0000000001 +y_0=0 +datum=NAD83 +units=us-ft +no_defs +ellps=GRS80 +towgs84=0,0,0')

# Google API - Samuel's (perhaps replace later?)
register_google(key = "AIzaSyD0R44_L_mFAxczh8cgTedmPDQRAmC2FBc")

m <- get_map("New York", zoom = 10, scale = 4, maptype = "roadmap")

library(ggrepel)
ggmap(m) +
  stat_density2d(aes(x=STOP_LOCATION_X, y=STOP_LOCATION_Y, fill = ..level.., alpha = ..level..),
               size = 2, bins = 10, data = as.data.frame(test_latlong),
               geom = "polygon") +
  geom_density2d(data = as.data.frame(test_latlong), aes(x=STOP_LOCATION_X, y=STOP_LOCATION_Y), size = 0.1)

# ggmap(m) + geom_point(aes(x=STOP_LOCATION_X, y=STOP_LOCATION_Y), data = as.data.frame(test_latlong))

manhattan <- get_map("Manhattan, New York", zoom = 12, scale = 4, maptype = "roadmap")

ggmap(manhattan) +
  stat_density2d(aes(x=STOP_LOCATION_X, y=STOP_LOCATION_Y, fill = ..level.., alpha = ..level..),
               size = 2, bins = 10, data = as.data.frame(test_latlong),
               geom = "polygon") +
  geom_density2d(data = as.data.frame(test_latlong), aes(x=STOP_LOCATION_X, y=STOP_LOCATION_Y), size = 0.1) +
  scale_alpha_continuous(range=c(0.1, 0.6))

library(plotly)

## 
## Attaching package: 'plotly'

## The following object is masked from 'package:ggmap':
## 
##     wind

## The following object is masked from 'package:ggplot2':
## 
##     last_plot

## The following object is masked from 'package:stats':
## 
##     filter

## The following object is masked from 'package:graphics':
## 
##     layout

# edit_r_environ(MAPBOX = your_unique_key)
Sys.setenv("MAPBOX_TOKEN" = "pk.eyJ1Ijoic2xlZTE5IiwiYSI6ImNrNDZmbWl1aTBqcmgzZW8xeDJzcTBsMXEifQ.OvJvDwdVBfB02NaJVLq7Fw")

library(mapdeck)

## 
## Attaching package: 'mapdeck'

## The following objects are masked from 'package:plotly':
## 
##     add_sf, add_text

key <- 'pk.eyJ1Ijoic2xlZTE5IiwiYSI6ImNrNDZmbWl1aTBqcmgzZW8xeDJzcTBsMXEifQ.OvJvDwdVBfB02NaJVLq7Fw'    ## put your own token here

test_latlong <- SQF1718[!is.na(SQF1718$STOP_LOCATION_X),]
test_latlong <- test_latlong[!is.na(test_latlong$STOP_LOCATION_Y),]

library(plotly)
library(sf)

## Linking to GEOS 3.7.2, GDAL 2.4.2, PROJ 5.2.0

# creating data as sf object
plot_locations <- test_latlong
coordinates(plot_locations) <- ~STOP_LOCATION_X + STOP_LOCATION_Y

proj4string(plot_locations) <- CRS('+init=epsg:2263 +proj=lcc +lat_1=41.03333333333333 +lat_2=40.66666666666666 +lat_0=40.16666666666666 +lon_0=-74 +x_0=300000.0000000001 +y_0=0 +datum=NAD83 +units=us-ft +no_defs +ellps=GRS80 +towgs84=0,0,0')

plot_locations <- plot_locations %>% spTransform(CRS("+init=epsg:4326"))

plot_locations_sp <- st_as_sf(plot_locations, coords = c("STOP_LOCATION_X", "STOP_LOCATION_Y"), crs = utm18nCRS)



# dividing the sf object into "Seven Major Felony Offenses" (NYC)
plot_locations_sf <- test_latlong %>%
  filter(SUSPECT_ARREST_OFFENSE %in% c("GRAND LARCENY", "ROBBERY", "BURGLARY", "ASSAULT", "RAPE", "TERRORISM", "GRAND LARCENY AUTO", "AUTO STRIPPING", "MURDER"))

coordinates(plot_locations_sf) <- ~STOP_LOCATION_X + STOP_LOCATION_Y

proj4string(plot_locations_sf) <- CRS('+init=epsg:2263 +proj=lcc +lat_1=41.03333333333333 +lat_2=40.66666666666666 +lat_0=40.16666666666666 +lon_0=-74 +x_0=300000.0000000001 +y_0=0 +datum=NAD83 +units=us-ft +no_defs +ellps=GRS80 +towgs84=0,0,0')

plot_locations_sf <- plot_locations_sf %>%
  spTransform(CRS("+init=epsg:4326"))

plot_locations_sf <- st_as_sf(plot_locations_sf, coords = c("STOP_LOCATION_X", "STOP_LOCATION_Y"), crs = utm18nCRS)

p <- plot_locations_sp %>%
  plot_mapbox(mode = 'scattermapbox', hoverinfo='name') %>%
  layout(font = list(color='white'),
         plot_bgcolor = '#191A1A', paper_bgcolor = '#191A1A',
         mapbox = list(style = 'dark'))

mapdeck(token = "pk.eyJ1Ijoic2xlZTE5IiwiYSI6ImNrNDZmbWl1aTBqcmgzZW8xeDJzcTBsMXEifQ.OvJvDwdVBfB02NaJVLq7Fw", style = mapdeck_style("dark"), pitch = 45) %>%
  add_screengrid(data = plot_locations_sp,
                 weight = "weight",
                 layer_id = "screengrid_layer",
                 cell_size = 6,
                 opacity = 0.4,
                 colour_range = colourvalues::colour_values(1:6, palette = "plasma")) %>%
  add_hexagon(data = plot_locations_sf,
              layer_id = "hex_layer",
              radius = 6
              )

## Registered S3 method overwritten by 'jsonify':
##   method     from    
##   print.json jsonlite

The bottom is a shiny coding that I will / published on my Shiny server

library(shiny)
library(shinydashboard)

## 
## Attaching package: 'shinydashboard'

## The following object is masked from 'package:graphics':
## 
##     box

ui <- dashboardPage(
    dashboardHeader(title = "New York Stop and Frisk")
    , dashboardSidebar(
        actionButton(inputId = "roads", label = "roads")
    )
    , dashboardBody(
        mapdeckOutput(outputId = "map")
    )
)

server <- function(input, output) {
    
    ## initialise a map
    output$map <- renderMapdeck({
        mapdeck()
    })
    
    ## use an observer to add and remove layers
    observeEvent({input$roads},{
        
        if ( input$roads %% 2 == 1 ) {
            
            mapdeck_update(map_id = "map") %>%
                add_path(
                    data = roads
                    , layer_id = "myRoads"
                    , stroke_colour = "RIGHT_LOC"
                    , update_view = TRUE
                    )
        } else {
          mapdeck_update(map_id = "map") %>%
                clear_path(layer_id = "myRoads")
        }
    })
}

shinyApp(ui, server)

## 
## Listening on http://127.0.0.1:5438

Project Overview (http://cs109hubway.github.io/classp/) - Motivation Example:

The Stop, Question and Frisk program is a practice, utilized by New York City Police Department (NYPD), of temporarily detaining, questioning, and at times searching civilians on the street for weapons and other contrabands (needs reivision). It is also called “Terry Stop” due to the Supreme Coourt ruling of Terry v. Ohio (1968). [Whether the rule prevents crime] - https://crim.sas.upenn.edu/fact-check/does-stop-and-frisk-reduce-crime

Seven Major Felony Offenses

http://www.nyc.gov/html/nypd/downloads/pdf/analysis_and_planning/seven_major_felony_offenses_historical_data.pdf

# Citywide Seven Major Felony Offenses 2000-2018 (Excel)
url <- "https://www1.nyc.gov/assets/nypd/downloads/excel/analysis_and_planning/historical-crime-data/seven-major-felony-offenses-by-precinct-2000-2018.xls"
destfile <- "seven-major-felony-offenses-by-precinct-2000-2018.xls" 
curl::curl_download(url, destfile)

• Background

• Objectives / Purpose
  • Is there an inherent bias in policing by precinct
  • Are police officers effective in identifying suspected crime (network?)
  • Geographic / race (contrasting)
• Apporach / Methodology
• Analysis / Findings

• Prediction

Comparison of Pre-, Bloomberg Era, and Post-Bloomberg Pre-Bloomberg –> maybe 2001 or 2002 stat? Michael Bloomberg (2002-2013) –> peak at 2011 Post-Bloomberg (2014-)

Priyanka: - models to see what factors actually mattered the most to see if there were discriminatory practices in their choices